THINKING ABOUT COMPETITIVE BALANCE

by

Allen R. Sanderson and John J. Siegfried

Working Paper No. 03-W18

September 2003

DEPARTMENT OF ECONOMICSVANDERBILT UNIVERSITY

NASHVILLE, TN 37235

www.vanderbilt.edu/econ

1

Thinking About Competitive Balance

Allen R. Sanderson, University of Chicago, and John J. Siegfried, Vanderbilt

University1

INTRODUCTION

Within the last decade, the business of American professional team sports has

seen contentious strikes and lockouts, escalation of players’ salaries, multi-million dollar

television and cable-rights contracts, increased melding of sports and entertainment, new

expansion teams, bitter franchise relocations, and an explosion in the construction of

publicly-financed facilities. In an era in which inequalities of wealth and opportunity are

constantly at the forefront of public-policy discussions in the United States, in the world

of professional sports many high-profile economic events also seem to turn on issues of

inequality: between owners and players over the distribution of economic rents; among

owners over the distribution of gate receipts, broadcast revenues, and talent; and between

mayors, taxpayers, owners and league officials over figurative level playing fields with

regard to the provision of state-of-the-art venues.

With regard to baseball, the re-emergence of the New York Yankees as a dynasty

in the post-strike period of the last decade – four World Series championships in a five-

year period (1996-2000), winning 116 games in 1998, and payrolls that dwarfed those of

most opponents – contributed as much as anything to the perception of an injurious

competitive imbalance problem. The Yankees’ loss in the 2001 World Series, the 2002

four-year collective bargaining agreement between Major League Baseball (MLB) and

the players’ union , which included measures designed to produce more balance, and/or

the first-ever all-wild-card World Series in 2002 (with the two teams’ combined payrolls

2

equaling the Yankees’) will likely not dampen enthusiasm for the topic among fans in

bars or on sports talk radio, by media commentators or by sports economists and

statisticians doing their analyses.

Simon Rottenberg (1956) long ago noted that “the nature of the industry is such

that competitors must be of approximately equal ‘size’ if any are to be successful; this

seems to be a unique attribute of professional competitive sports.” (242)2 While the

absolute quality of play influences demand and absolute investments in training are

socially efficient (Lazear and Rosen, 1981), the relative aspects of demand and quality of

competition also loom large in sports. In cases when consumer demand depends to a

large extent on inter-team competition and rivalry, the necessary interactions across

“firms” (i.e., teams) define the special nature of sports. Contests among poorly matched

competitors would eventually cause fan interest to wane and industry revenues to fall.

But potential “arms races” – or “rat races” – are possible, and maybe even inevitable

(Akerlof, 1976).3

In the 1990s and early 21st century, while some social scientists wrung their hands

over apparent widening inequality of income in the United States and between developed

and third-world nations, sports economists, commentators, fans and owners (at least

among the also-rans) lamented the perceived widening inequality of wealth and

championships among teams, especially the large-market, high-revenue, and/or high-

payroll teams versus their country-cousin kin in baseball.4 The distribution of playing

talent, team revenues and salary expenditures, and competitive balance dominate sports

and sports-business conversations whenever a team with the highest payroll, or an owner

with the deepest pockets or a decidedly different utility function, wins a championship; or

3

whenever a franchise extracts a new publicly-funded ballpark from its community on the

threat of relocating; or when an apparent dynasty emerges, such as the New York

Yankees in recent years or another NL-East division title for the Atlanta Braves.5

Even before the baseball season began in 1999, the league adopted the mantra,

reiterated upon seemingly every public occasion (with the uncritical assistance of the

sports media) that few of baseball’s franchises actually stood a chance of being in the

World Series in the fall. In a Wall Street Journal column that year, San Diego Padres’

chairman John Moores, whose team was swept 4-0 by the Yankees in the 1998 World

Series and then traded away much of its top talent, echoed these sentiments:

“The Major League Baseball season began yesterday, and fans everywhereare hoping their team will make it to the World Series. But the sad fact isthat many teams are chronically uncompetitive. Perhaps 12 of the 30Major League teams have any possibility of reaching postseason play, andfewer still have a realistic hope of winning a pennant. Unless baseballchanges the way it does business, it risks seeing its fans drift away, tiredof their teams’ futility.”6

Also in 1999, baseball commissioner Allan H. “Bud” Selig convened a panel of

well-known individuals7 to study the impact of revenue disparities on competitive

balance. It produced a lengthy report, The Commissioner's Blue Ribbon Report on

Baseball Economics (Levin et al, 2000), that noted large and growing revenue disparities,

which, in turn, affected balance.8 In addition to more quantitative theoretical and

empirical measures of competitive balance (see below), the Blue Ribbon panel also

defined competitive balance qualitatively:

"In the context of baseball, proper competitive balance should beunderstood to exist when there are no clubs chronically weak becauseof MLB's financial structural features. Proper competitive balancewill not exist until every well-run club has a regularly recurringhope of reaching postseason play." (page 5)

4

Broadcaster Bob Costas chimed in with similar laments (Costas, 2000). Major

League Baseball conducted a national poll of 1,000 fans in late 2001 that purported to

indicate that competitive imbalance was a serious problem in the minds of 75 percent of

respondents; 42 percent of them indicated they would lose interest in the game were more

teams not to have a realistic chance of winning. Summarizing the results, Sandy

Alderson, MLB's Executive Vice President, said: "We have a competitive-balance

problem. This is something the average fan cares about. They don't care if owners are

losing money. They do care if it translates into negative consequences for their teams."9

Although certainly not unique to this particular period or sport, the complaint of

woeful imbalance has become more common in the last few years, with the lack of

significant revenue sharing or a firm payroll cap in Major League Baseball (MLB)

relative to the National Football League (NFL) identified as the culprits. Increased player

freedoms, through which an owner could hire the best players and, at least in the short

run, buy a championship, is seen as an accomplice. Apart from the obvious owners’

interest in limiting bidding for players, and the equally obvious interest among players in

that not being allowed to occur, payroll caps, salary caps, luxury taxes, increased revenue

sharing and restructured draft systems are touted as ways to constrain competition and

thus improve competitive balance among teams.10 League restrictions on both the

geographical relocations of teams and the mobility of players across teams, in addition to

having more self-serving purposes, also affect balance. A more radical proposal was the

commissioner's recent threat to contract baseball by as many as four teams.

In the sections below we attempt to lay out what one might mean by competitive

balance, review the theoretical and empirical scholarship and popular contributions with

5

regard to its various dimensions, describe the natural forces and considerations, as well as

institutional rules and regulations that contribute to observed distributions of playing

performances. We also compare at various junctures the situation in baseball versus

other sports leagues, including college athletics, and individual sports.

COMPETITIVE BALANCE IN THEORY AND PRACTICE

Every sport and sports league has had to confront the fundamental issue of

relative strengths among competitors. There has not been a uniform, one-size-fits-all

approach or set of rules to resolve this problem. Inasmuch as uncertainty of outcome is a

key component of fan demand, wide disparities in inputs, and thus in likely outcomes, are

seen as inimical to the long-term health and financial viability of the individual

enterprise. How to handle weak teams or inferior opponents – to prevent lower quality

competitors from free-riding on higher quality rivals – can be as much or more of a

problem as dealing with perennially strong ones, because there is at least some interest in

seeing the very best individual performers and teams.11

Boxing segments fighters into weight classes and employs rankings and ladders to

create bouts with equally matched opponents. Auto racing, track competitions and

swimming use qualifying times to ensure competitive fields. Tennis produces seedings,

based on previous performances, in the expectation that the strong will play the strong in

later-round matches. “Claiming races” in thoroughbred racing is a mechanism designed

to have horses of approximately equal ability entered into the same event. Except for the

occasional novelty or promotion, women do not compete against men. Periodic structural

changes or modifications in the rules of play, such as elimination of the center-jump and

6

adoption of a shot clock in basketball, or altering the height of the mound or changing the

effective strike zone in baseball, have been used to tilt the playing field to achieve a

certain objective and to prevent some competitors from exploiting particular advantages

or decreasing fan interest in the contests. Another example is restrictions on athletes’ use

of performance-enhancing substances.

The theory of competitive balance in team sports was first developed by

Rottenberg (1956).12 While not a direct contribution to the sports-economics literature,

concepts in Ronald Coase’s “The Problem of Social Cost” (1960) have been applied to

reserve clauses, reverse-order draft systems, and player free agency as they affect, or fail

to influence, playing strengths and the reallocation of player resources.13 Since that time,

economists have contributed rigorous theoretical and empirical work on various aspects

of competitive balance (Daly and Moore (1981), Gerald Scully (1989 and 1995), Quirk

and Fort (1992, 1995, and 1999), Zimbalist (1992), Dobson and Goddard (2002, Chapter

3)), including formal measures of balance within a league (such as the dispersion of won-

lost percentages and the concentration of championships); momentum (or serial

correlation); league rules with regard free agency, restricting entry and expansion; cross-

subsidization schemes such as reserve and draft systems, caps and revenue sharing; and

the connection between payroll and performance.

Important comparisons between baseball before and after free agency, before and

after the 1994-95 players’ strike, and with other professional team sports, have appeared

in the literature in recent years (Horowitz (1997), La Croix and Kawaura (1999), Depken

(1999), Vrooman (2000), Eckard (2001A, 2001B), and Zimbalist (2001)). In its first

three years of existence, the Journal of Sports Economics published five separate articles

7

plus an entire special issue on competitive balance.14 For English soccer, Szymanski

(2001) has shown that the growing financial disparity among clubs has had no impact on

imbalance. How closely payroll and/or market size correlate with winning, including, of

course, the determination of the causal relationship, is arguably one of the most important

questions about competitive balance. It is also essential to evaluate the relationship

between market size and team payroll, which often inaccurately is assumed to be tight.

More popular accounts from sports commentators, reporters and fans may lack the

economists’ technical sophistication, but they substitute the appeal earlier applied by

some to pornography – they can’t define balance formally or precisely, but they

recognize it when they see it, and they are convinced they don’t see it now. The Blue

Ribbon Report (BRR) noted that recently baseball teams with the largest payrolls

dominate post-season play. Its authors observed that all eight playoff teams in1999 were

among the top 10 in payroll, the Dodgers and Orioles being the only exception. The BRR

reported (page i) that: “From 1995 through 1999, a total of 158 postseason games were

played. . . . During this five-year period, no club from [the bottom half of the payroll

distribution] won a Division Series or League Championship Game.”15 It also noted the

positive relationship, in theory and practice, between winning percentage and market

size, concluding that large-market teams are likely to emerge disproportionately at the top

of league standings and in the distribution of championships.16

Sports columnists have devised their own competitive balance criteria. Allen

Barra (2002A) has proposed two complementary competitive imbalance measures with

regard to post-season play: the number of different franchises that make the playoffs, and

the number of different teams that play for a championship. Comparing MLB to the NFL

8

and the NBA, Barra noted that in spite of fewer available playoff spots, over the last two

decades 20 franchises have appeared in the World Series, compared with 19 in the Super

Bowl and 15 in the NBA finals. Since 1996, the period of heightened concern about

imbalance in baseball, approximately the same number of different teams have appeared

in the World Series as in championship series in the other two leagues. In Clearing The

Bases (2002B, Chapter 13), Barra adds a third metric: the number of teams in baseball

that finish the regular season with better than a .600 or worse than a .400 winning

percentage, arguable standards of domination and futility, respectively, that he notes have

been met by fewer and fewer teams over time.17

To the extent that fan demand is a function of a possible competitive outcome on

any particular occasion as opposed to an entire season or in the hunt for a championship,

in baseball the premier team in any season loses about four games in 10, and generally

fares worse on the road than at home. In the 2000 season, for example, the World

Champion New York Yankees lost 74 games, almost 46 percent of their contests. They

swept only six of the 39 three-game regular season series they played that year; thus

“Yankee haters” had a reasonable probability of having their own team prevail on any

given night or any trip the Bronx Bombers made to town.

THE NATURAL DEMAND FOR AND SUPPLY OF BALANCE (AND IMBALANCE)

Apart from the constraints leagues place on competition to ensure balance (factors

discussed later), which may have the complementary effect of increasing and/or

redistributing revenues, there are also natural forces that influence the distribution of

outcomes. One obstacle to reducing inequality in sports leagues could be, paraphrasing

9

Pogo, that we may not only be content with the current imbalance but also actually prefer

it to the alternatives. Or, at a minimum, we are conflicted and willing to let natural (and

unnatural) forces and inertia, rather than explicit interventions, determine outcomes. In

economic and sporting walks of life, we have a preference for a positive correlation

between effort and reward. To reward the statistically better individual or team for its

prior achievements we tip some balances in its favor – playing more games at home,

higher seeds or a better lane, an extended playoff series rather than a single winner-take-

all contest. The more evenly matched two opponents are, the higher the probability that a

random element – a poor call by an official, a bad bounce, a key injury, or pure luck –

will determine the outcome.18 Thus, the premise that the demand for games is greater,

ceteris paribus, the greater the degree of uncertainty conflicts with our sense of justice

that the better team win. Luck is the way we account for the success of people and teams

we don't like, but it is not a factor that we generally want to determine our income

distributions in society or our champions in sporting contests.

On the other hand, we have strong identifications with and sympathies for the true

underdog. We want David to knock off Goliath, at least on occasion (unless, of course,

Goliath plays for “us” – Chicago was quite content with the Bulls’ domination in the

1990s, and Yankee fans have few quibbles with the alleged imbalance in baseball).

Nowhere is this better exemplified than in popular sports movies that feed off imbalance.

Films such as "Rocky," "Hoosiers," "The Mighty Ducks," and "Major League" cater to

these instincts. (And, after all, the play was called "Damn Yankees," not "Damn Cubs.")

As much as one may loathe a bully, selling tickets to an event in which he has some

10

chance of being upset is marketable. Dynasties, storied franchises, such as the Yankees,

Celtics, Packers and Red Wings are not without their advantages in terms of fan interest.

The world is replete with examples of healthy inequality more extreme than the

current levels in baseball. State lotteries are popular despite daunting odds. While just

25 of the nation's 400 graduate schools grant a third of all new Ph.D.s each year, the

"industry" maintains diversity and vitality while competing implicitly head to head.

Other comparable concentrations and inequalities abound – live theater, world-class

symphonies, and first-rate art museums are not evenly distributed across the landscape.

In sports, virtually all teams win at least a fourth of their games and few win more

than two-thirds of the time. Victories – and losses – are not inevitable. In leagues with

30 teams, the probability of winning the ultimate prize – a World Series or Super Bowl –

with equal distributions of talent in any given year is 0.033; thus a team or city could

expect to garner a championship about once a generation. The difference between that

periodicity and a “dynasty” can be as little as a factor of three, in that in the latter

instance a team may win a championship once a decade instead of once per generation.

Sports' fans' memories are selective and the rate at which they fade appears to be

small. That a team last won a Super Bowl or World Series more than a decade ago may

seem like yesterday. (Many Chicago fans think of the Bears as a championship team,

even though their only Super Bowl win occurred in 1986.) Stadiums and arenas are

replete with banners from past conquests, in most cases many years removed. Fans don

sweatshirts and caps that evoke past memories as much as current realities. That, coupled

with our basic "hope springs eternal" (or "wait until next year") spirit, fueled with

optimism about the most recent draft choice or free-agent acquisition, buoys the soul –

11

and ticket sales. Scully's (1995) empirical validation of serial correlation in sports and

his admonition that fans should be patient reinforces our natural instincts and outlooks.

In a society that confronts substantial inequality in its daily experiences, the current level

of imbalance in baseball may not be intolerable.

In addition to these many factors contributing to the observed inequality of

outcomes, whatever the metric, there are several other possible explanations for

competitive team imbalance. The following examples are but a few such considerations.

Differences in Population and Preferences

The demand for beachwear in Ft. Lauderdale dramatically exceeds the demand for

swimsuits in Buffalo and also in St. Augustine, yet no one seems to worry about bikini

imbalances. Indeed, we would probably be puzzled if such sales in Ft. Lauderdale did

not exceed combined receipts in Buffalo and St. Augustine by a large margin. Although

the populations of Buffalo and Ft. Lauderdale are similar, the per capita demand for

bathing suits is greater in Florida than in upstate New York, presumably because of

greater utility in use. And, although St. Augustine and Ft. Lauderdale share the same

climate and accessibility to beaches, the population of Ft. Lauderdale is many times

larger than St. Augustine. Geographic differences and preferences exist with regard to

types of foodstuffs consumed, automobiles driven, and television programs watched.

So, too, for winning sporting contests. Residents in some locations may be

willing to pay more to have a more successful local team (e.g., per capita willingness and

ability to support a winning ice hockey team undoubtedly is greater in Ontario than in

Florida), especially if there are fewer other recreational, entertainment and/or cultural

12

amenities close at hand. Population disparities across areas hosting teams can create

differences in aggregate willingness and ability to pay even when individual customers in

the various host cities have identical tastes.

These differences could be equalized if teams in sports leagues were free to move

to areas where the marginal revenue per win is higher than their initial location. The

resulting competition among teams in the same league within a metropolitan area would

dilute the incremental revenues earned by the original incumbent, and reduce financial

disparities among the teams. The collection of Australian Rules football teams in

Melbourne, the concentration of baseball teams in Tokyo, and the density of premier-

league soccer teams in London illustrates the possibility.

Team movements that would help to equalize marginal revenues do not occur,

however, because each of the professional men's team sports leagues in North America

exercises a form of collective control over member team movements. Incumbents in the

larger cities or those cities where fans are willing to pay more for winning are loath to

share their revenues with immigrants from smaller communities or from locations where

success on the playing field is less important to the residents. They are protected from

incursions either by league constitutional provisions protecting their "home territory," or

by their ability to form coalitions sufficient to prevent other teams from moving into their

host community. In short, the competitive imbalance that emanates from the monopoly

control of home territories by incumbents arises from the conduct of the leagues' member

teams themselves because they are unwilling or unable to introduce competition into

areas where favored incumbent teams earn considerable economic rents. Revenue

13

disparities among franchises in professional baseball would diminish if the Montreal

Expos were permitted to move to northern New Jersey.19

These disparities, and even the “large market” versus “small market” distinction,

would disappear if expansion occurred within an existing cartel league, a rival league

formed, and/or some judicial action broke up the cartel (as the resulting smaller leagues,

in search of new markets, spawned new franchises).20 New York City is considered a

large market and Kansas City a small one in part because the ratio of teams to population

is so dissimilar. Migration of existing franchises and/or the creation of new ones located

in larger metropolitan areas would equalize these ratios in sports leagues much as the way

in which retail establishments and other social amenities equalize in more traditional,

competitive markets, like fast-food outlets and swimsuit retailing.

Willingness to Act on Differences in Fan Tastes

In addition to different preferences for winning, fans who live in different areas

may differ in their willingness to act on those preferences (Porter, 1992). The more

fickle are fans, the more their willingness to buy tickets depends on the local team's on-

field success, the greater is the marginal revenue from the local team's winning additional

contests, and the greater is the incentive for a team to expend resources to secure more

highly skilled players. A profit-maximizing league could exploit fickleness by

strengthening teams located where fans are less loyal and, in turn, weakening teams

where fans will turn out regardless of the success of the local franchise. This would not

bode well for the Cubs and Steelers. If a league did this, however, the distribution of

14

playing talent could be inefficient because it is configured in response to fans' willingness

to act on their preferences rather than on the basis of the preferences themselves.

Moreover, in addition to caring about the on-field success of the home team, fans

often have preferences regarding dominance that are independent of which team

dominates. That is, the competitive balance distribution itself is a public good. Everyone

must live with the same overall distribution, and individuals may have strong preferences

about the shape of that distribution. Some may prefer imbalance, even sufficient to create

dynasties, whether they love them or love to hate them. Others may prefer a league in

which almost all teams win about half of their contests.

While fans may have views on what the general distribution of competitive

balance should look like, these views may fluctuate over time. As the real income levels

of sports consumers have risen over the second half of the twentieth century (Siegfried

and Peterson, 2000), and the typical fan has moved further up an increasingly disperse

income distribution, fan preferences toward the overall distribution of competitive

balance conceivably could migrate toward more imbalance in game outcomes as well.

Differences in Player Tastes

Differences in team playing skills can arise even if there are no differences in the

population base of team territories, in fan fickleness, or in fan preferences toward home

team winning or dynasties because players also have preferences. Players may have a

preference for living in certain areas (e.g., Florida or Southern California) and be willing

to sacrifice part of their salary in order to do so.21 Or they may see more lucrative

endorsement opportunities in areas with relatively stronger media markets. Ken Griffey,

15

Jr. apparently accepted a lower salary than he could have earned elsewhere to sign a

contract to play in his hometown of Cincinnati; the Mariners were attractive to John

Olerud because of family ties in Seattle – and Lou Pinella engineered a move from

Seattle to Florida to manage closer to home. Wayne Gretsky precipitated his trade to Los

Angeles to accommodate his wife's acting career. And on the other side, Steve Francis

announced he was never going to live in Vancouver, and so stimulated the (then)

Vancouver Grizzlies to trade him to the Houston Rockets.

Network economies can also affect the distribution of playing talent. Players may

accept a lower salary to be on a team with greater odds of winning a championship,

thereby further enhancing the talent of the contender. Ray Bourque approved a trade

from the Boston Bruins to the already powerful Colorado Avalanche late in his career to

boost his chance of playing on a Stanley Cup championship team, a goal he achieved in

2001. Greg Maddox’s self-relocation from the Cubs to Braves is another example.

Consider, in the extreme, how players might distribute themselves across teams if

their salaries were zero. Would they join teams so as to balance playing talent? College

football and basketball may provide some guidance as to what might happen because

players' compensation is relatively constant across universities. There the attractive new

prospects gravitate toward perennial winners in order to enjoy greater prospects for on-

field success, to play with more talented teammates, and to gain more media exposure.

The Tradeoff between Winning and Uncertainty

Competitive balance is thought to affect attendance of fans through its influence

on winning and fans' response to winning. It is well established that home attendance

16

rises when a team wins more games or matches, and declines when it loses.22 Winning

teams also attract more fans when they play on the road. The Yankees attract a crowd

when they visit other American League cities, whether it is because fans love to hate a

winner, wish to see marquee players, or the Yankees have a following in other locations.

As the mobility of our population has increased in the post World War II years, fans of

particular teams more frequently reside outside their home territory. Similarly, “super-

station” broadcasts have created national Braves and Cubs fans. While that phenomenon

does not affect home attendance much, it can affect away attendance and television

ratings. The extent to which owners of stronger teams take these factors into account

depends on league rules for sharing gate and television revenues. The Detroit Red Wings

draw huge crowds on the road because they are an elite team of established (read: older)

talented players, and because former residents of Detroit maintain their affinity for the

team and attend games that the Red Wings play on the road. This generates incremental

revenue for other National Hockey League teams, but does not affect incentives facing

the Red Wings management because in the NHL visiting teams do not share in gate

receipts and there is little national television revenue.

Competitive balance may also affect attendance negatively at games among teams

that are relatively weak because fans view them as "out of the running" for a

championship (as opposed to any individual contest on any given day). If this occurs

when the revenue sharing rules are confined largely to home and visiting teams, the

owners of the higher revenue teams may ignore the external impact of imbalance on the

league's overall revenues until it reaches the point where the integrity of the overall

competition is called into question and fans abandon the sport altogether. The addition of

17

wild-card teams and smaller divisions or conferences, which both increase the number of

teams eligible for the playoffs, are innovations designed to retain fans across more cities

longer into the regular season by increasing the uncertainty of ultimate outcomes.

Character of the Events Themselves

In the 1980s and 1990s, responding to fan demand driven in part by higher

incomes, professional sports franchises repositioned their product by increasing the

emphasis on complementary, non-contest services and amenities, blurring further the

distinction between sports and entertainment. (The short-lived XFL may have overshot

that moving target.) Although the Dallas Cowboys Cheerleaders pre-date the 1980s, they

are a vivid representation of the movement. New stadiums that include upscale

restaurants, batting cages and other amusements for children, museums, Jumbotron

scoreboards with instant-replay and promotions, fireworks, and even a swimming pool

(in Phoenix), reduce the relative importance of the game in the overall recreational

package. Intermission entertainment at basketball, ice hockey, and football games is now

standard fare, including contests for fans, mini shows by well-known musical artists,

trampoline groups, and scoreboard video clips. Even the Montreal Forum has abandoned

the traditional fare of just a Zamboni machine doing the ice between periods of NHL

contests, and the New York Islanders have replaced traditional ice-maintenance workers

(read: middle-aged, overweight males in parkas) with navel-baring young women in

Lycra outfits.23 Luxury boxes and exclusive access areas increase the value of "being

seen" at the game relative to seeing the game.

18

As the relative importance of the game itself diminishes in the entertainment

package, competitive balance becomes a less important determinant of demand. So long

as the tendency to broaden the entertainment experience is similar across locations of

differing revenue potential, however, this phenomenon, like revenue sharing, should not

have more than a modest effect on competitive balance.

Complementary Economic Theory

Traditional economic arguments for what prevents one owner from amassing an

all-star line-up in an effort to win every contest and the championship turn on notions of

self-interest and the inevitable diminishing marginal returns to, and the increasing

marginal cost of, victories. Hence some natural mechanisms constrain the extent of

imbalance in most leagues. That assumes, however, that there are both effective ways to

blunt possible negative externalities, and that owners, deep down, are profit maximizers.

Common practices within sports leagues to ensure some semblance of a "level

playing field" with regard to the distribution of talent across teams – reverse-order draft

systems, various attempts to constrain players' salaries, revenue sharing – are also at odds

with how economic theory generally views "peer effects" and the optimal sorting of

workers. Where spillovers are significant, high ability workers are more valuable to

other high quality workers, and workers should be more homogeneously sorted.24 The

"assignment problem" involves how to sort heterogeneous units into groups so that output

is maximized.25 Whether in general production processes, law firms, or marriages, higher

ability workers, colleagues or spouses are more productive when grouped with other high

ability people. Because similar quality workers are more productive when sorted into the

19

same firm, a firm with higher quality workers may not be willing to employ lower quality

workers even if those workers would work for less pay.

The general application of this sorting principle in sports implies, for example,

that an all-star shortstop's productivity is higher if he is paired with an all-star second

baseman. Moreover, on-the-field traits carry over into the clubhouse and social settings,

where discipline, motivation, attitude and joint-monitoring can be important as well. If

this proposition is true, practiced in sports it would lead to inequality in the distribution of

talent – some teams would have good players, other teams poor ones, and competitive

imbalance would emerge. (Within baseball organizations, the farm system may serve as

a sorting mechanism, and across-team trades and free-agent signings may represent, in

part, attempts to capture potential peer-effect gains.) In some sense, then, sports leagues

may be fighting an uphill battle in trying to stem the tide of nature and market forces

pushing toward imbalance.

INSTITUTIONAL ARRANGEMENTS AND COMPETITIVE BALANCE

Many institutions or off-the-field rules of the game are negotiated every time a

Collective Bargaining Agreement (CBA) is renewed. These changes may affect the

degree of competitive balance in any professional sports league.

Payroll Caps and Luxury Taxes

Payroll caps, and so-called "luxury taxes" on payrolls, create an incentive for

owners of teams in higher revenue locations to hire less talent than they would in the

absence of these constraints. In the extreme, a binding ceiling on total payroll limits the

20

amount of talent a high revenue team can accumulate. As a by-product, a firm payroll

cap also increases the profits of high revenue teams. Its impact on competitive balance

depends on the extent to which the cap is below the free-market payroll level of the

highest payroll team that also dominates on the field (read: Yankees).

One danger of payroll caps is that they may be porous (e.g., the NBA cap that

includes a well-known loophole – the “Larry Bird” exemption – designed to preserve

team unity, or the NFL, in which for 2002 accounting conventions permitted virtually

every team in the league to be over the cap), and they create a temptation for violating the

rules (e.g., the Minnesota Timberwolves paid a large fine for paying Joe Smith "off the

books" in excess of the cap). A system in which cheaters are more successful than those

who play by the rules may be even less inviting than one in which teams fortunate

enough to own rights to high revenue areas are more successful. Moreover, payroll (and

salary) caps do not extend to complementary inputs, so successful coaches – Jon Gruden,

Dusty Baker, Phil Jackson – can command a sizable sum for managing a team on which

the total payroll, and even individual salaries, may be frozen.

A less drastic version of the payroll cap is what has come to be known in

professional sports as a "luxury tax" on payrolls, first enacted in baseball for three

seasons (1997-99) and reinstituted in the 2002-2006 agreement as a 17.5 percent penalty

on payroll amounts exceeding $117 million for the 2003 season.26 The rationale is that

fielding a highly paid team is a "luxury” for one owner that imposes negative

externalities on other franchises. This makes sense if the tax becomes effective at the

point where incremental talent on the high revenue team creates a league-wide net

negative impact that might be ignored by the owner of the high revenue team because

21

under league revenue sharing rules he or she bears little of the cost of an "over-

accumulation" of talent. If the tax rate accurately reflects this internal "externality," it

creates an incentive for the high revenue team owner to balance his or her gain against

the cost to third parties. The trick, of course, is to impose the tax rate at the proper

payroll level, and to fix the rate such that it internalizes the externality.

To be accepted as “fair,” luxury tax revenues usually compensate those who bear

the burden of the externality. This tax in MLB is, indeed, structured properly to achieve

these goals, although no one knows whether the threshold payroll, the tax rate, and the

beneficiaries of the redistribution are properly identified.

Salary Caps

The NBA is the only U.S. men's professional sports league currently using

individual player salary caps to control team expenditures. Maximum salaries are based

on seniority in the league. Salary caps emerged from the NBA's collective bargaining

with the players’ union in 1998 and early 1999, at least partly in reaction to the league's

leaky team payroll cap.

Individual salary caps limit a team's payroll to the product of the roster size times

the cap for the most senior players, not a significant constraint. Individual salary caps

based on seniority are unlikely to have much of an impact on competitive balance

because a high revenue team can sign a complete team at the highest allowed salary,

thereby accumulating an entire team of the most desirable players in the league. This is

not likely to happen, however, because the most expensive team one can buy is also an

old team. A team can assemble a more competitive roster paying less than the maximum.

22

If league rules constrain salaries, free agent players' choices of which team to join

will turn more on their personal preferences, including desirable places to live and

prospects for endorsements and a championship. Individual salary caps are likely to

increase competitive imbalance because they encourage players to rely more on their

preference for joining a winning team than on differences among salary offers. Salary

caps will, however, limit the payroll of the high revenue teams, because they are the

teams that would have bid above-cap salaries to acquire the more talented players.

Individual player salary caps probably help the highest revenue teams increase both their

profits and their playing talent.

Revenue Sharing27

Revenue sharing reduces the financial incentive of each franchise to acquire more

talent, because the payoff to winning is constrained by the share paid to other franchises.

Sharing revenues that are sensitive to playing success blunts the incentive to win for all

franchises – those in low revenue potential locations as well as those in New York.

Because the demand for winning may vary across communities with the intensity

of competition for playing talent (fans in some communities find it more satisfying to win

when there is a dogfight for talented players), revenue sharing can affect competitive

balance. If fans in high revenue potential locations are relatively more sensitive to

winning when competition for players (from all teams) is more intense, then increased

revenue sharing will improve competitive balance, because the incentive to acquire better

players will be muted further for high revenue potential teams. Of course, if the fans in

high revenue potential locations are less sensitive to winning when competition for

23

players is more intense, then expanded revenue sharing will exacerbate competitive

imbalance. Contrary to popular belief, the effect of revenue sharing on competitive

balance, while likely modest, could go in either direction.

If revenue sharing blunts the incentive for all teams to bid aggressively for

talented players, thereby muting salary differentials between more and less talented

players, non-pecuniary considerations will loom larger in free agents' decisions among

competing offers. If players value the opportunity to play on championship contenders

for reasons beyond financial rewards, increased revenue sharing thus could lead to

greater competitive imbalance, a result likely to surprise many people.

The premier men's professional team sports leagues engage in a variety of

revenue sharing arrangements. The NFL is the most socialistic, sharing revenues from its

huge national television contract and merchandise sales (almost) equally, and gate

receipts 66-34 to the home and visiting teams.

The two baseball leagues modestly share local revenues. Ice hockey and

basketball share the fewest revenues; in both the NHL and NBA the home team retains

all gate receipts. A relatively new wrinkle in baseball is “dynamic pricing” (called

“variable pricing” in the press), long a feature in other industries, including airlines and

hotels, through which ticket prices vary not only by quality of the seat location but also

by day of the week, month, and, now, opponent. In the 2002 season, for example, the

Colorado Rockies charged higher prices when the Yankees and Giants played in Denver

than when the Brewers visited. More baseball clubs implemented such pricing for the

2003 season. The Chicago Cubs’ 2003 schedule listed Wednesday and Thursday day

games in April against the Expos and Padres as “value” dates; summer weekend games

24

versus the Cardinals, Yankees and White Sox are “prime home games,” with prices to

match. To the extent that Kansas City can charge fans more for a popular opponent like

the Yankees, this represents implicit revenue sharing that narrow inequality in team

revenues across a league and thereby promotes more competitive balance.

Number of Teams and Relocation Restrictions

Just a few days after the 2001 World Series, Major League Baseball announced a

plan "contract" in order to ameliorate the effects of competitive balance. Missing from

the analysis was a theory relating the number of teams to competitive imbalance.

Presumably the theory is that lopping off the bottom of any distribution tightens it

up. Of course, lopping off the top would also tighten it up. And while it is not

reasonable to suggest that MLB contract by eliminating the New York Yankees and

Atlanta Braves, although that would certainly have an effect on competitive balance not

unlike disbanding the Montreal Expos and the Minnesota Twins, an even better

alternative would be to let the Expos and Twins move into the New York metropolitan

area. A movement from low revenue potential to high revenue potential locations would

certainly improve competitive balance, the objective identified by MLB, more than

would just eliminating two cellar-dwellers. Indeed, the core of the competitive imbalance

problem is the differences in population and tastes for sports across metropolitan areas

coupled with each league's artificial restriction on team movements. Let the teams move

to wherever they like and the differences in revenue potential will dissipate and

competitive balance will improve.28 Yes, high revenue potential teams will be less

25

profitable, but if protecting the profits of teams fortunate enough to have the largest

demand for their local monopoly is the goal, then explain the contraction on that basis.

To the extent that a competitive balance targeted policy also affects profitability

by restraining team payrolls, it may also affect the number of teams. Limitations on team

payrolls make lower revenue potential locations viable prospects for joining the league

and, in turn, tempt owners, who pocket league initiation fees, to expand the number of

marginal franchises and, thus, by their own conduct, reduce competitive balance.

Reverse-Order Player Drafts29

Reverse-order “rookie” drafts, long a staple in many leagues, may have a modest

effect on competitive balance, but in baseball it is indeed modest. The difference among

teams in drafting is at most one player per year: The worst team in a 30-team league

drafts first and 31st, and the best team drafts 30th; thus beyond the first round there is no

draft advantage to the worst team relative to the best. The net gain in terms of balance is

simply one player, which in baseball cannot usually make a substantial difference.30

To the extent that teams are free to transfer player contracts, the initial allocation

of property rights – whether to the worst team in the league or the best team in the league

– implies no difference in the ultimate allocation of player resources, and, hence,

competitive balance. Unless there are constraints on player sales, a player new to the

league who is expected to generate greater marginal revenue product at a franchise with

high revenue potential than at one with low revenue potential will end up playing for the

high revenue potential franchise, either by initially signing with it, or because the lower

revenue potential franchise sells his contract to the high revenue potential franchise.

26

A considerable number of empirical tests of the Coase Theorem were spawned by

the change in property rights over players that occurred during the 1970s, when players

with a certain amount of seniority gained the rights to "free agency."31 Most of these

studies concluded that the transfer of property rights from teams to players did not affect

the distribution of playing talent, although it did raise player compensation levels.

Other Revenue Enhancements

Luxury box revenues, stadium naming rights, field advertising, and other local

revenue sources can affect competitive balance if they are not proportional to other

revenues among teams. If stadium naming rights and luxury box revenues, for example,

constitute a higher ratio to gate receipts for higher revenue potential teams, these added

revenues will increase the disparity among teams and may exacerbate competitive

imbalance. On the other hand, naming rights and luxury box revenue are not likely to be

sensitive to team performance. Although a team in a high revenue potential location may

enjoy more of these peripheral revenues than one in a lower revenue potential area,

unless the revenue is linked to improved team performance on the field, there is no

incentive for the team to spend the revenue enhancement on improving its talent level.

Structure of the Competition

Much in the same way that financial arrangements can affect balance among

competitors, so too can the nature and structure of the contests themselves.32 How pole

positions in auto racing, lane assignments in swimming, brackets in tennis tournaments,

or starting gates in thoroughbred races are determined can influence outcomes. In team

27

sports, comparable issues turn on how home field, court or ice is established for contests,

and the length of the series.

Lengthening a series reduces the probability that the weaker opponent will win;

increasing the number of playoff rounds and the percentage of teams eligible for a

championship reduces the chances that the best team will capture the championship

(Sanderson, 2002). Phil Rogers, baseball analyst for the Chicago Tribune, notes that with

the advent of divisional play in 1969, teams with the best regular-season record over 162

games have fared relatively poorly in the playoffs. Since 1990, only the 1998 Yankees

won a World Series after finishing with the best record during the season, and only five

times out of 12 have teams with the most wins during the season even made it to the

Series. Since 1969, only eight teams with the best season record have won the World

Series.33 In the three years 2000-2002, the team with the better record has won only

twice in 17 series (Rogers, 2002). Reducing the number of playoff teams, say, by

reverting to only one AL and one NL division in baseball, with the winner of each league

meeting in the World Series, would ensure that the teams with the best regular-season

record met for the championship, but would entail serious tradeoffs – a loss of excitement

late in the season and forgone playoff revenues, for example.

In baseball, about a quarter of teams make the playoffs, compared to about half

the teams in the other three professional men’s team sport leagues. If baseball were to

emulate ice hockey, football, or basketball with regard to the number of playoff-eligible

teams and playoff rounds, the probability that the team with the highest payroll would

appear in, or win, the World Series would be lower, thus dampening its owner’s incentive

to spend money for talent, and probably producing more balance across teams.34

28

A final “macro” structural change that would replace the weakest teams with

stronger ones, long familiar in soccer (European football), is the operation of open

leagues, with promotion and relegation. Closed leagues, when combined with reverse-

order drafts and revenue sharing, tend to reward failure and punish success, whereas open

leagues reward success and punish failure. (See Dobson and Goddard, 2002; Noll, 2002;

and Rosen and Sanderson, 2001.)

WHY IS ALL OF THIS SIMPLY A BASEBALL PROBLEM?

In Major League Baseball over the last 20 years 35, a free-agency period, 20

different teams have played in a World Series, and 14 different teams have won at least

one championship. For the previous 20 years, largely pre-free-agency, 14 different teams

played in a World Series and 10 won titles. From 1981-2001, the Yankees appeared in a

World Series six times, winning four; the Atlanta Braves won once in five tries; the

Oakland A's and St. Louis Cardinals each played in three Series and won once; no other

team appeared more than twice. For the previous 20-year period, the Yankees won four

times and lost three; the Dodgers won twice and lost four times; and the Orioles and Reds

each won two times in five tries. The record shows that in spite of the Yankees’ recent

success, free agency and high payrolls have not increased the concentration of

championships. The two largest markets – New York and Los Angeles, which host two

teams each – were represented in World Series competitions 16 times between 1961 and

1980 but only 11 times in the last 20 years.36

Still, the Blue Ribbon Report, the commissioner’s office, commentators, and

everyday fans insist that Major League Baseball has an untenable competitive imbalance

29

problem that has worsened since 1995. Proposed remedies include caps on payrolls

and/or salaries, more revenue sharing, and even contraction. Needless to say, players and

their association (MLBPA) disagree – with both the facts and proposed solutions that

would directly or indirectly reduce individual or collective paychecks.

Proposals to bring more balance to baseball often rely on comparisons to

professional football and basketball, with their salary and payroll restrictions and/or

greater revenue sharing. A more valuable “what if” benchmark may be Division I college

football and men’s basketball, where there is no free agency and "payrolls" are

approximately level.37 Outcomes there do not support the conclusion that restrictions on

payrolls and other conditions of play would necessarily produce more balance.

For example, in the last 20 years (1981-2000), 17 of the more than 100 Division I

football programs have been ranked in the AP (Associated Press) Top 20 nationally for at

least 10 years, Nebraska achieving that distinction 19 times. In basketball, 10 programs,

led by Duke, North Carolina and Kentucky, account for more than half of all Final Four

appearances over the period. When "payrolls" (i.e., tuition, fees, room and board) are

approximately the same, non-pecuniary considerations loom larger. A coach, the team's

schedule, an institution's track record (i.e., its reputation) with regard to winning and

preparing its athletes for professional careers, and other local amenities take on more

importance. (At the professional level, flattening payrolls would benefit teams in major

markets such as New York and Los Angeles relative to areas with fewer pecuniary

opportunities and non-monetary amenities.)

Why is there so little outrage over the domination of selected athletic programs at

the collegiate level? Why are we more than willing to countenance more inequality year

30

after year in college athletics, other professional sports leagues, and in individual sports,

but not in baseball? A few college football programs – Miami, Florida, Florida State,

Nebraska, Oklahoma – play in top bowl games most every January; Duke, North

Carolina, Kansas, Kentucky, Arizona, and UCLA are the men's basketball counterparts.

In NCAA baseball, LSU won five championships in the 1990s, thus exhibiting more

dominance than the Yankees in a larger field of competitors. Arkansas has a lock on

indoor and outdoor men's NCAA track and field titles. Texas and Stanford dominate

men's and women's collegiate swimming and diving. Stanford and Georgia "own" men's

and women's collegiate tennis. Iowa has won 19 of the last 20 wrestling championships.

North Carolina dominates women’s soccer. In 2002, the Connecticut Huskies women’s

basketball team, winners of four of the last eight national championships, set an NCAA

record for women’s basketball by winning 55 consecutive games.38 Yet we hear few

laments about these imbalances. Why is baseball singled out?

More unbalanced achievements prevail in many professional individual sports

than in baseball. We revel in Lance Armstrong's "fourpeat" in the Tour de France. Ty

Murray has won 8 of the last 12 years' all-around Professional Rodeo Cowboys

Association world championships. The concentrations of victories of Tiger Woods and

Annika Sorenstam in golf and the Williams sisters in tennis become stories closely

followed by the media and fans alike.

CONCLUSIONS

Of the many contemporary controversies in baseball – ticket prices, owners’

financial losses, contentious labor-management disputes, the level and rate of growth of

players’ salaries, among others – the one that has arguably received the most attention

31

recently is the alleged lack of competitive balance. It was the principal focus of the

commissioner’s Blue Ribbon Report, front and center with regard to proposed provisions

of the 2002 Collective Bargaining Agreement, and is a constant theme with the press,

public, and economists writing about baseball.

Does competitive imbalance in baseball deserve so much attention? How much of

all this is real rather than random? How much confuses a trend with a cycle? How much

of the observed outcomes are immutable to policy prescriptions, private or public? Is

there anything there beyond four Yankees’ titles in a five-year period, and could this be

akin to the occasional string of four consecutive heads arising from repeated coin tosses?

Would baseball be better off if it looked more like the NFL, or would football trade some

of its parity for other dimensions of demand and voluntarily move more toward baseball

with respect to competitive imbalance? Some of this is likely to be revealed – and

modified – by the time baseball owners and players sit down at the bargaining table in

2006. Until then, questions about relationships between payrolls and performance, city

size and winning, uncertainty of outcomes and the level of demand, and the role of player

and team mobility in competitive balance will remain a fertile field of dreams for

econometricians and commentators alike to hoe.

32

REFERENCES

Akerlof, George. 1976. “The Economics of Caste and of the Rat Race and Other Woeful Tales.” Quarterly Journal of Economics. 90 (4). 599-617.

Baade, Robert A. and Allen R. Sanderson. 1997. “Cities Under Siege,” in WallaceHendricks (ed), Advances in the Economics of Sport, Volume 2. Stamford, CT: JAIPress. Pages 77-114.

Barra, Allen. 2002A. “Imbalanced Logic.” The Wall Street Journal. August 30, p. W5.

Barra, Allen. 2002B. Clearing The Bases. New York, NY: St. Martin’s Press.

Becker, Gary S. 1973. "A Theory of Marriage." Journal of Political Economy, Part I. 81: 813-846.

Bloom, Matt. 1999. “The Performance Effects of Pay Dispersion on Individuals and Organizations.” Academy of Management Journal. 42, No. 1. [February]: 25-40.

Coase, Ronald. 1960. “The Problem of Social Cost.” Journal of Law and Economics. 3 [October]: 1-44.

Costas, Bob. 2000. Fair Ball: A Fan’s Case for Baseball. New York, NY: Broadway Books.

Daly, George and William J. Moore. 1981. “Externalities, Property Rights and the Allocation of Resources in Major League Baseball.” Economic Inquiry. 19 [January]: 77-94.

Depken, Craig A., II. 1999. "Free-Agency and the Competitiveness of Major League Baseball." Review of Industrial Organization. Vol. 14: 205-217.

Dobson, Stephen and John Goddard. 2001. The Economics of Football. New York, NY: Cambridge University Press.

Eckard, E. Woodrow. 2001A. "Free Agency, Competitive Balance, and Diminishing Returns to Pennant Contention." Economic Inquiry. Vol. 39, No. 3, July: 430-443.

Eckard, E. Woodrow. 2001B. “Baseball’s Blue Ribbon Report: Solutions in Search of a Problem.” Journal of Sports Economics. Volume 2, Number 3. August: 213-227.

El-Hodiri, Mohamed and James Quirk. 1971. “An Economic Model of a Professional Sports League.” Journal of Political Economy. 79 [November/December]: 1302-19.

Fort, Rodney. 1992. “Pay and Performance: Is the Field of Dreams Barren?” In Paul

33

M. Sommers (ed), Diamonds are Forever. Washington, DC: The BrookingsInstitution. 134-157.

Fort, Rodney and James Quirk. 1995. “Cross-Subsidization, Incentives, and Outcomesin Professional Team Sports Leagues.” Journal of Economic Literature. XXXIII[September]: 1265-99.

Hausman, Jerry A. and Gregory K. Leonard. 1997. “Superstars in the NationalBasketball Association: Economic Value and Policy,” Journal of Labor Economics.Vol. 15, No. 4 [October], 586-624.

Horowitz, Ira. 1997. "The Increasing Competitive Balance in Major League Baseball." Review of Industrial Organization. Vol. 12: 373-387.

Hylan, T.R., Lage, M.J. and Treglia, M. 1996. “The Coase Theorem, Free Agency, andMajor League Baseball: A Panel Study of Pitcher Mobility from 1961 to 1992.”Southern Economic Journal. Vol. 62 [April]: 1029-42.

Frank, Robert H. and Phillip J. Cook. 1995. The Winner-Take-All Society. New York, NY: The Free Press.

Guryan, Jonathan. 2001. "Estimating Peer Effects in the Workplace: Evidence from Random Pairings In Professional Golf Tournaments." Mimeo.

Humphreys, Brad R. 2002. “Alternative Measures of Competitive Balance in Sports Leagues.” Journal of Sports Economics. Volume 3, Number 2; May: 133:148.

Hunt, Joseph W. and Kenneth A. Lewis. 1976. “Dominance, Recontracting and theReserve Clause: Major League Baseball,” American Economic Review. Vol. 66, 936-43.

Kesenne, Stefan. 2000. “Revenue Sharing and Competitive Balance in Professional Team Sports.” Journal of Sports Economics. Volume 1, Number 1. February: 56-65.

Koopmans, Tjalling C. and Beckman, Martin. 1957. "Assignment Problems and the Location of Economic Activities." Econometrica. 25: 53-76.

Kremer, Michael. 1993. "The O-Ring Theory of Economic Development." Quarterly Journal of Economics. 108: 551-575.

La Croix, Sumner J. and Kawaura, Akihiko. 1999. "Rule Changes and Competitive Balance in Japanese Baseball." Economic Inquiry. Vol. 37, No. 2, April: 353-368.

Lazear, Edward and Sherwin Rosen. 1981. “Rank-order tournaments as optimum labor contracts. Journal of Political Economy. Volume 89, page 841-864.

34

Levin, R.C., Mitchell, G.J., Volcker, P.A. and Will, G.F. 2000. The Report of theIndependent Members of the Commissioner’s Blue Ribbon Panel on BaseballEconomics. New York: Major League Baseball.

Marburger, Daniel. 1997. "Gate Revenue Sharing and Luxury Taxes in Professional Sports. Contemporary Economic Policy. Vol. XV, no. 2. 114-123.

Neale, Walter. C. 1964. “The Peculiar Economics of Professional Sports: AContribution to the Theory of the Firm in Sporting Competition.” Quarterly Journalof Economics. Vol. 78 [February]: 1-14.

Noll, Roger G. 2002. “The Economics of Promotion and Relegation in Sports Leagues,” Journal of Sports Economics, Vol. 3, No. 2, May: 169-203.

Porter, Phillip K. 1992. “The Role of the Fan in Professional Baseball,” in Paul M. Sommers (ed)., Diamonds are Forever: The Business of Baseball. Washington, DC: The Brookings Institution Press, 63-76.

Quirk, James and Rodney Fort. 1992. Pay Dirt: The Business of Professional Team Sports. Princeton, N.J.: Princeton University Press.

Quirk, James and Rodney Fort. 1999. Hard Ball: The Abuse of Power in Pro Team Sports. Princeton, N.J.: Princeton University Press.

Rogers, Phil. 2002. “Wins don’t translate to playoffs.” Chicago Tribune. October 13, sec 3, page 2.

Rosen, Sherwin and Allen Sanderson. 2001. “Labour Markets in Professional Sports.” The Economic Journal. Vol. 111, No. 469 [February]: F47-F68.

Ross, Stephen F. 1989. “Monopoly Sports Leagues.” Minnesota Law Review, 649.

Ross, Stephen F. 2001. “Light, Less-Filling, It’s Blue-Ribbon.” Unpublished.

Rottenberg, Simon. 1956. “The Baseball Players’ Labor Market.” Journal of Political Economy. 64 [June]: 242-58.

Rottenberg, Simon. 2000. “Resource Allocation and Income Distribution in Professional Team Sports. Journal of Sports Economics. Volume 1, Number 1. February: 11-20.

Saint-Paul, Gilles. 2001. "On the Distribution of Income and Worker Assignment underIntrafirm Spillovers, with an Application to Ideals and Networks. Journal of PoliticalEconomy,109: 1-35.

Sanderson, Allen R. 2002. “The Many Dimensions of Competitive Balance.” Journal of Sports Economics. Volume 3, Number 2. May: 204-228.

35

Sanderson, Allen R. and John J. Siegfried. 1997. “The Implications of Athlete Freedom to Contract: Lessons from North America.” Economic Affairs. 17 [September]: 7-12.

Schmidt, Martin B. and David J. Berri. 2001. “Competitive Balance and Attendance:The Case of Major League Baseball.” Journal of Sports Economics. Volume 2,Number 2. May: 145-167.

Schmidt, Martin B. and David J. Berri. 2002. “Competitive Balance and Market Size inMajor League Baseball: A Response to Baseball’s Blue Ribbon Panel.” Review ofIndustrial Organization, 21: 41-54.

Scully, Gerald W. 1989. The Business of Major League Baseball. Chicago: The University of Chicago Press.

Scully, Gerald W. 1995. The Market Structure of Sports. Chicago: The University of Chicago Press.

Siegfried, John J. 1995. “Sports Player Drafts and Reserve Systems.” Cato Journal. Vol. 14, No. 3 [Winter]: 443-452.

Szymanski, Stefan. 2001. “Income Inequality, Competitive Balance and theAttractiveness of Team Sports: Some Evidence and a Natural Experiment fromEnglish Soccer,” The Economic Journal. Vol. 111, No. 469 [February]: F69-F84.

Szymanski, Stefan. 2003. “The Economic Design of Sporting Contests: A Review.” Journal of Economic Literature. (Forthcoming)

Utt, Joshua and Rodney Fort. 2002. “Pitfalls to Measuring Competitive Balance with Gini Coefficients.” Journal of Sports Economics. Vol 3, No. 4 November

Vrooman, John. 1995. "A General Theory of Sports Leagues." Southern Economic Journal. [April]: 971-990.

Vrooman, John. 2000. "The Economics of American Sports Leagues." Scottish Journal of Political Economy. 17 [September]: 364-398.

Walker, Sam. 2000. “Hockey’s Vision in Lycra.” The Wall Street Journal. November 1. Page W4.

Whitney, James D. 1993. "Bidding Till Bankrupt: Destructive Competition in Professional Team Sports." Economic Inquiry. Vol. XXXI, January: 100-115.

Zimbalist, Andrew. 1992A. Baseball and Billions. New York, NY: Basic Books.

36

Zimbalist, Andrew. 1992B. “Salaries and Performance: Beyond the Scully Model.” InPaul M. Sommers (ed), Diamonds Are Forever. Washington, DC: The BrookingsInstitution. 109-133.

Zimbalist, Andrew. 2001. “Competitive Balance in Major League Baseball.” The Milken Institute Review. Vol. 3, No. 1. [First Quarter]: 54-64.

Zimbalist, Andrew, Guest Editor. 2002. “Symposium: Competitive Balance in Sports Leagues.” Journal of Sports Economics. Volume 3, Number 2. May.

Zimbalist, Andrew. 2003. May The Best Team Win. Washington, DC. The Brookings Institution Press.

Allen R. Sanderson is associate chair and senior lecturer in economics at theUniversity of Chicago.

John J. Siegfried is Professor of Economics at Vanderbilt University andSecretary-Treasurer of the American Economic Association.

37

NOTES

1 We are indebted to Brad Humphreys for thoughtful comments on an earlier draft and toRod Fort, Steve Ross and Andy Zimbalist for reactions to the version presented at theconference.2 See also Rottenberg, 2000.3 See also Whitney (1993).4 Protests of the New York Yankees’ payroll in 1999 included Kansas City fans turningtheir backs when the Yankees came up to bat, and then filing out of Kauffman Stadium inthe third inning. Similar greetings accompanied the high-salaried Alex Rodriquez in2001, when he returned to Seattle now as a member of the Texas Rangers.5 The Yankees have won four World Series in the last seven years and 26 in all, thoughprior to 1996, they had not been in a World Series since 1981 and had not won one since1978. That club’s on-field fortunes were cited in Rottenberg’s seminal article to illustratethe impact, of lack thereof, of the reserve system on the distribution of playing talent inbaseball. However, that the Yankees have won 26 of the 38 World Series they haveplayed in – a statistically significant 68.4 percent – may attest more to their grit and guilethan it does to their payroll. (From 1991 to the present, the Atlanta Braves have wontheir division title 11 consecutive years. However, their only World Series win, in fiveappearances in the decade, came in 1995.)6 “Bring Competition Back to Baseball,” The Wall Street Journal, April 5, 1999. Thatsame week, pre-tournament coverage of the Masters golf tournament included a similarrefrain: “So while 96 players will tee off, far fewer have a realistic chance to win. ‘Howmany have a chance of winning?’ said Fred Couples, who won the Masters in 1992, andwho tied for second last year. ‘Fifty. But realistically, I’d say 10 or 12 guys.” [The NewYork Times, April 8, 1999, C21.] (Three days later, Jose Maria Olazabal, who was notamong the favorites, though he had won in Augusta in 1994, won the 1999 tournament.)7 Economist and Yale University President Richard Levin, former Senator GeorgeMitchell, former Federal Reserve Board Chairman Paul Volcker, and political pundit andauthor George Will.8 Comments on and criticism of the BRR and its conclusions include papers by Ross(2001), Eckard (2001B), and Schmidt and Berri (2002).9 Chicago Tribune, December 18, 2001. Section 4, pages 1 & 6.10 A century ago, of course, the reserve clause was the principal mechanism employed byowners to allegedly ensure balance; outright collusion that restrained bidding for freeagents was the tool of choice in baseball twenty years ago; organizing new leagues, suchas Major League Soccer (MLS) and the WNBA, as “single entity” structures is a morerecent manifestation of similar goals. Limits on roster size are often thought to enhancecompetitive balance by preventing some teams from stockpiling talent; however, there islittle literature on this subject and scant empirical support for that contention.11 The New York Yankees have led the league in attendance – at home and on the road –in recent years. The Montreal Expos and Tampa Bay Devil Rays do not draw well athome, nor do they fill ballparks for their away games. During the Jordan era, the Bullsplayed to sold-out arenas around the NBA, the only team to do so. See Hausman andLeonard (1997) for an analysis of the impact of superstars Larry Bird, Magic Johnson and

38

Michael Jordan on attendance and television viewership – and revenues – in the NBA,and the extent of free riding by other franchises.12 El-Hodiri and Quirk (1971) also treat this issue in their classic article.13 For example, see Siegfried (1995) and Hylan et al (1996). Another early contributionwas from Walter Neale (1964).14 Volume 1, Number 1, February 2000, articles by Rottenberg; Kesenne; Volume 2,Number 2, May 2001, article by Schmidt and Berri; Volume 2, Number 3, August 2001,article by Eckard; Volume 3, Number 3, May 2002 (entire issue); Volume 3, Number 4,November 2002, article by Utt and Fort; and Volume 4, Number 1, February 2003,contains commentaries on competitive balance by Eckhard and Humphreys.15 On the same page of the BRR, it is claimed that: “From 1995 through 1999, only threeclubs achieved profitability.” This assertion has been questioned by many observers.16 Schmidt and Berri (2002) do not find a link between market size and competitivebalance. They also contend that the causality between these two factors is ambiguous. Inanother recent variation on the inequality theme, applying the work of Frank and Cook(1995), Bloom (1999) investigates intra-team pay dispersion and finds, holding totalpayroll constant, a negative relationship between the degree of inequality in a team’spayroll and individual (and thus team) on-field performance. Nevertheless, in baseball atleast, the within-team salary inequality, between the minimum and maximum or the rangebetween the median and the highest salary, continues to widen.17 In 2000, for the first year in MLB history, no team finished above .600 or below .400.18Questionable officiating during the NFL playoffs arguably determined Super Bowlopponents in both 2002 and 2003.19 Or even to Puerto Rico, where the Expos played a fourth of their 2003 “home”schedule.20 See Ross (1989) and Baade and Sanderson (1997) for arguments and proposals forbreaking up existing leagues into smaller entities.21 Sports commentators have occasionally referred to San Diego’s climate, beauty andlifestyle as creating a “Padre discount” in that team’s ability to sign players for a lowersalary than would be true for, say, the Detroit Tigers.22Szymanski’s review of the literature (2003) shows that empirical support for thecorrelation between winning and demand exists, but that it is weaker than usuallyassumed. In his 2001 Economic Journal article on English soccer, Syzmanski (2001)finds that match attendance appears to be unrelated to competitive balance.23 Sam Walker, “Hockey’s Vision in Lycra,” The Wall Street Journal, November 1, 2002,W4.24 The authors are indebted to Todd Kendall for pointing out this literature andcomplementary arguments.25 See Koopmans and Beckman (1957), Becker (1973), Kremer (1993), Saint-Paul (2001)and Guryan (2001).26 See Zimbalist (2003) for lengthy discussions of the labor-management processsurrounding the 2002 Collective Bargaining Agreement.27 See Marburger (1997).28 Of course, the purchased right to operate in “exclusive territories” is commonplace andexpected in franchise operations and many other commercial areas, and generally mayserve efficiency purposes, not just in professional sports.

39

29 Additional draft dimensions include supplemental drafts, similar to expansion drafts, atthe conclusion of regular seasons, and, for MLB, the inclusion of non-U.S. players in theannual amateur draft. We do not treat either of these variations in our paper.30 Arguably, the draft could have more of an impact in the NBA, where one player couldconstitutes 20 percent of a starting line-up and there is more agreement about a player’spotential, than in football or baseball, which have larger rosters and predictions ofperformance are less reliable.31 See Hylan et al (1996) and Krautmann and Oppenheimer (1994).32 See Szymanski (2003) for a discussion of the design of contests, including the optimalnumber of entrants, structure of rewards, and amount of revenue sharing and balance.33 Remarkably, in 17 of the Yankees’ 26 championships they won the final game on theWorld Series on the road.34 Historically, the potential home-field advantage in the World Series – when a seriesgoes the full seven games – has been rotated each year between the two leagues. Thecommissioner’s recent proposal, which owners and players approved, lets whicheverleague wins the annual All-Star game have that edge in 2003 and 2004, could affectbalance. The team with the home-field advantage actually lost the World Series in 21 ofthe 30 years from 1955-84, including Game 7 a total of 12 times. However, in the last 22World Series played, the team that has been home for Games 1 & 2 (and potentially 6 &7), has prevailed 18 times (and 15 of the last 17), including the last 8 decided by aseventh game (no visiting team has won Game 7 since the Pittsburgh Pirates in 1979).35 1981-2001; there was no World Series played in 1994 because of the players' strike.36 The top four teams in the earlier period accounted for 23 of 40 available World Seriesslots. In the most recent 20 years the four most represented teams comprise only 17 of 40possible appearances. In the National Football League, 19 different franchises haveplayed in a Super Bowl in the last 20 years, and 12 different teams have won at leastonce. The San Francisco 49ers have recorded the most wins (4) and the Buffalo Bills lostall four Super Bowl games in which they played. Five franchises account for more thanhalf of the appearances and two-thirds of the wins. In the last 10 years alone, 15 differentteams have played in the Super Bowl; none of the six divisional champions from the2000 season repeated in 2001. Over the last 20 years in the NBA, 14 different teamshave played in a championship series and seven of them have won at least once; with sixwins each, the Lakers and Bulls account for 12 of the 20 trophies. In the NHL over thesame period, 18 teams played in the Stanley Cup Finals, with Edmonton making the mostappearances (6, winning 5).37 There are, of course, some additional differences between the NBA or NFL and theircollege counterparts, including a university being unable to threaten to relocate if a newstadium isn’t forthcoming, the non-profit nature of higher education and, unlike forprofessional franchises, the more complex, multi-dimensional goals of educationalinstitutions.38 The men’s basketball record for consecutive victories is 88, set by UCLA in the early1970s (1971-74); Oklahoma won 47 straight football games in the 1950s.